While researchers have acknowledged the importance of community environment on physical activity, few systematic investigations of the relationship between community physical and social characteristics and physical activity have been undertaken. Drawing upon theoretical and empirical developments in urban sociology, urban planning, transportation, and health services research, the proposed analyses assess the hypothesis that the built environment and social environment of urban community affects physical activity over and above individual background. We propose a research project that investigates the designs of the built environment and multiple facets of community social, economic, demographic, and cultural environment in their influences on physical activity. Using multiple data sources including the Census data, CIS-based measures and two existing survey data, this research will address the following questions:1) How does the design of the built environment affects physical activity over and above individual social and demographic background?; 2) How do community-level social factors such as trust, safety, physical disorder, feelings of community morale affect individual likelihood of physical activity independently of the effects of the built environment?; 3) Whether and how do the built environment and social environment interact to impact physical activity? For example, does the effect of neighborhood walk ability vary according to neighborhood SES or other aspects of social environment such as trust? 4) What are the ecological relationship between the built environment and social environment at the community level?; and 5) How are the above relationships sensitive to geographic aggregation levels of community? What are the most appropriate aggregation levels relevant to physical activity? [unreadable] [unreadable] [unreadable]